Transient Hemodynamics Prediction Using an Efficient Octree-Based Deep Learning Model

Maul N, Zinn K, Wagner F, Thies M, Rohleder M, Pfaff L, Kowarschik M, Birkhold A, Maier A (2023)

Publication Type: Conference contribution

Publication year: 2023

Journal

Publisher: Springer Science and Business Media Deutschland GmbH

Book Volume: 13939 LNCS

Pages Range: 183-194

Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Event location: San Carlos de Bariloche

ISBN: 9783031340475

DOI: 10.1007/978-3-031-34048-2_15

Abstract

Patient-specific hemodynamics assessment could support diagnosis and treatment of neurovascular diseases. Currently, conventional medical imaging modalities are not able to accurately acquire high-resolution hemodynamic information that would be required to assess complex neurovascular pathologies. Instead, computational fluid dynamics (CFD) simulations can be applied to tomographic reconstructions to obtain clinically relevant information. However, three-dimensional (3D) CFD simulations require enormous computational resources and simulation-related expert knowledge that are usually not available in clinical environments. Recently, deep-learning-based methods have been proposed as CFD surrogates to improve computational efficiency. Nevertheless, the prediction of high-resolution transient CFD simulations for complex vascular geometries poses a challenge to conventional deep learning models. In this work, we present an architecture that is tailored to predict high-resolution (spatial and temporal) velocity fields for complex synthetic vascular geometries. For this, an octree-based spatial discretization is combined with an implicit neural function representation to efficiently handle the prediction of the 3D velocity field for each time step. The presented method is evaluated for the task of cerebral hemodynamics prediction before and during the injection of contrast agent in the internal carotid artery (ICA). Compared to CFD simulations, the velocity field can be estimated with a mean absolute error of 0.024 m s ^{- 1}, whereas the run time reduces from several hours on a high-performance cluster to a few seconds on a consumer graphical processing unit.

Authors with CRIS profile

Involved external institutions

Siemens AG Germany (DE)

How to cite

APA:

Maul, N., Zinn, K., Wagner, F., Thies, M., Rohleder, M., Pfaff, L.,... Maier, A. (2023). Transient Hemodynamics Prediction Using an Efficient Octree-Based Deep Learning Model. In Alejandro Frangi, Marleen de Bruijne, Demian Wassermann, Nassir Navab (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 183-194). San Carlos de Bariloche, AR: Springer Science and Business Media Deutschland GmbH.

MLA:

Maul, Noah, et al. "Transient Hemodynamics Prediction Using an Efficient Octree-Based Deep Learning Model." Proceedings of the 28th International Conference on Information Processing in Medical Imaging, IPMI 2023, San Carlos de Bariloche Ed. Alejandro Frangi, Marleen de Bruijne, Demian Wassermann, Nassir Navab, Springer Science and Business Media Deutschland GmbH, 2023. 183-194.

BibTeX: Download